DESCRIPTION: (Applicant's Abstract) A variety of strategies have been developed to accomplish gene therapy for cancer. The strategy of mutation compensation attempts to achieve genetic correction of cancer cells by specifically targeting the genetic events etiologic of neoplastic progression. This has been previously attempted employing antisense methods of gene inhibition. Whereas phenotypic reversion has been demonstrated in several contexts using this method, problems related to antisense instability and delivery have limited clinical application of this approach. As an alternative, the applicant has developed a novel method to achieve specific ablation of relevant oncogenes. In this regard, the applicant has developed methods to accomplish intracellular expression of a single-chain antibody (sFv) directed against the oncoprotein. In this schema, the newly synthesized oncoprotein would be entrapped during synthesis, and thus unable to exert its transforming effect. As an initial target, the applicant has developed methods to achieve selective ablation of the oncoprotein erbB-2, which has been shown to be important in the pathogenesis of ovarian and other carcinomas. For the approach that the applicant proposes, anti-erbB-2 single-chain immunoglobulin (sFv) genes were constructed to direct expression of intracellular anti-erbB-2 antibodies. Preliminary studies showed that expression of an endoplasmic reticulum (ER) form of the anti-erbB-2 sFv resulted in a profound down-regulation of cell surface erbB-2 in an erbB-2 over-expressing ovarian carcinoma cell line. In addition, expression of the intracellular antibody resulted in a direct cytocidal effect in erbB-2 overexpressing tumor cells. Further analysis demonstrated that this cytocidal effect was on the basis of induction of apoptosis. Thus, mislocalization of the erbB-2 oncoprotein in the ER of tumor cells accomplished a novel means of inducing cellular apoptosis. Importantly, expression of the ER anti-erbB-2 sFv in non-overexpressing cells did not induce cell death or apoptosis. Thus, this form of genetic intervention is capable of achieving a targeted killing of erbB-2 overexpressing tumor cells. This finding has been translated into a gene therapy approach for ovarian carcinoma whereby a recombinant adenovirus encoding the anti-erbB-2 sFv could be shown to accomplish in vivo tumor cell killings in a mouse xenograft model of human ovarian carcinoma. Based upon these results, the applicant will further explore the mechanisms and utility of the anti-oncogene sFv technique. In this regard, studies will be undertaken to evaluate the molecular determinants of sFv-mediated apoptosis after oncoprotein mislocalization. In addition, other ovarian cancer knock-out targets will be addressed to determine their role in neoplastic progression and the therapeutic utility of achieving their functional knock-out in the context of ovarian carcinoma. Finally, in vivo studies will be carried out to develop the methods for translation of these strategies into a gene therapy approach for ovarian carcinoma.